The present invention relates to a method and a device for improving the quality of a cellular telecommunications network.
In general, optimizing a cellular network consists in estimating whether the global quality of the network, under certain assumptions concerning its parameters, satisfies certain predetermined criteria, and in varying the parameters or the criteria in order to tend towards optimized performance.
In the present state of the art, the criteria used may be classified amongst: coverage criteria; interference criteria; capacity criteria; and non-coverage criteria.
These criteria present the major drawback of needing to be defined relative to performance thresholds.
For example, a coverage criterion is meaningful only in association with a minimum required field level on which it is possible to base optimization.
These thresholds may be constant and predefined in the optimization tool, or they may be adjustable by the user.
For example, the document “Towards automatic cell planning” by R. Kapp Rawnsley and S. Hurley; Department of Computer Science, Cardiff University, Cardiff CF24 3XF, published in 2000 under the number 0-7803-6465-5/00 describes an optimization method in which use is made of two optimization criteria that are associated with thresholds that need to be defined by the user, namely:                a coverage criterion defined relative to a signal-to-noise ratio threshold set at 14 decibels (dB); and        a capacity criterion defined relative to a traffic threshold set at 43 Erlangs.        
The document “Methodology for WiMax performance optimization using drive test data” by F. Alghamidi, I. Kostanic, the 8th International Conference on Computing, Communications Control Technologies, Orlando, Fla., Apr. 6-9, 2010 proposes a method of optimizing a cellular network in which a coverage threshold in terms of field level is calculated automatically on the basis of the bandwidth and on the basis of a noise factor of the receiver.
In any event, defining thresholds associated with optimization criteria is tricky and difficult to optimize, regardless of whether these thresholds are determined empirically or automatically.
Furthermore, performance criteria based on such minimum performance thresholds do not guarantee the global performance of the network. For example, if the minimum field level is found to be too low, because it has already been reached over the entire coverage of the cellular network, or too high since it cannot be reached, then optimization has no effect.
Consequently, the invention seeks to provide a method and a device for optimizing a cellular telecommunications network in which there is no need to define a threshold level.